Peripheral nerve traction injuries may occur after surgical care and can involve any of the lower extremity large peripheral nerves. In this review, the authors discuss injuries after knee or hip surgical intervention. The diagnosis, including electrodiagnostic studies, is time sensitive and also relies on a detailed history and physical examination. Successful prevention and treatment involve familiarity with risk and predisposing factors as well as prophylactic measures.
Perioperative peripheral nerve traction injury is a poorly understood known complication with multiple etiopathologic considerations.
Presentation may vary along the spectrum of sensory and motor nerve injury.
Nerve injury can occur along any part of the path traversed with bias of predisposing factors like medical comorbidity and female gender.
Nerve conduction and electromyogram (EMG) studies are invaluable for determining the presence of nerve injury, localization, completeness, recovery, and prognostication.
The authors examine potential nerve and plexus injury surrounding hip and knee surgery.
Diagnosis is established with thorough history and physical examination, along with the aid of testing modalities, such as magnetic resonance imaging/ultrasound and diagnostic testing, such as EMG/nerve conduction studies.
Careful assessment of patients, planned procedure, and risk factors are key to prevention.
NOTE: Please see preceding article entitled, “Perioperative Upper Extremity Peripheral Nerve Traction Injuries.”
Peripheral nerve traction injuries of the lower limb
Following Hip Surgery
Nerve injuries related to hip surgery may be the most-studied neurologic complication of joint surgery. The incidence of nerve damage following total hip arthroplasty (THA) varies from 0.09% to 7.6%, with incidences after primary THAs lower and after THA revision on the higher end. Regarding hip arthroscopy, the incidence of nerve injury is comparable; one aggregation of the results of 30 studies found an 1.7% prevalence. The exact cause of these postoperative peripheral nerve injuries are unknown 50% to 60% of the time time. Overall, most nerve palsies after THA have not been explored, with the assumption being that they are secondary to stretch injuries and would have a poor prognosis with exploration.
Farrell and colleagues looked specifically for motor nerve palsy in 27,000 cases of THA and found the incidence to be 0.17%. In a review of 3126 consecutive total hip replacements, the incidence of nerve injury in hip arthroplasty was 1.7% overall, 1.3% when looking at primary hip replacements, and as high as 3.6% in hip revisions. The same study also reported a much higher incidence for nerve injury in hip replacements for congenital hip dislocations and the developmental dysplasia of the hip (5.2%). Again the researchers do not differentiate the type of nerve injury that make up this category, admitting that the cause was unknown in up to 57% of cases who had a peripheral nerve injury. They do attribute 13 cases to be secondary to tension caused by the limb lengthening involved in some of the hip replacements. Considering only these cases of neuropathy after hip replacement, the incidence of peripheral nerve traction injuries (PPNTI) would then be only 0.4%.
These numbers show that hip revisions, congenital dislocations/dysplasia, and leg lengthening are 3 risk factors for nerve injury. Other studies have come to the same conclusions. Another risk factor that seems specific for hip arthroplasty is female sex. The surgical approach used in THA may also be a risk factor. Farrell’s results showed that the risk of nerve injury was significantly higher in patients operated on through a posterior approach than it was in those treated with an anterolateral or transtrochanteric approach.
The sciatic nerve accounts for more than 80% of hip surgery–associated PPNTI, most of which is the peroneal division, with an isolated injury to the tibial portion being rare. The next most commonly injured are the pudendal and femoral nerves. Typically, the reported ratio of pudendal to sciatic nerve injury is approximately 1:3 with the patient in the lateral position and 2:1 with the patient in the supine position. PPNTI of the superior and inferior gluteal as well as the obturator nerves are even more rare.
Sciatic nerve injury (L4–S3)
The association between sciatic neuropathy and hip arthroplasty is well established, with an incidence of approximately 1% to 2%. The incidence of sciatic nerve injury in acetabular fracture repair is approximately 2.9%. These diagnoses are clinical, and at least one study has shown that the prevalence of nerve changes seen with monitoring with transcranial motor and/or somatosensory evoked potentials is greater than what is clinically identified. In that study of 76 patients undergoing hip arthroscopy in the lateral position, 58% had intraoperative nerve dysfunction, whereas only 7% were diagnosed with a clinical nerve injury.
The common peroneal component of the sciatic nerve is not only more commonly affected than the tibial portion but, when both portions are affected, the peroneal injury is also usually more severe. Sudderland reported that, in 30% of the specimens that he has studied, the sciatic nerve exits the pelvis as 2 separate nerves. In such people, the tibial division will enter the gluteal region distal compared with the peroneal division. Also the peroneal component is the more lateral and superficial portion of the sciatic nerve; once separate, it is relatively more fixed between the sciatic notch and the fibular head. These anatomic reasons make the peroneal portion (or division) more vulnerable to injury, such as traction and compression from retractors.
Yet another proposed explanation is based on the morphologic differences between the tightly packed fascicles of the peroneal division and those of the tibial division, which have relatively more connective tissue. The abundant connective tissue, as in the tibial division, protects those nerves and makes them less vulnerable to transection or compression than nerves with tightly packed fascicles such as in the peroneal division. Also, because the peroneal division nerve bundles are larger, in accord with Laplace’s Law, they are at a higher risk of compression compared with tibial nerve bundles.
Another predisposing or risk factor to mention is patients who underwent hip fracture surgery, had preoperative traction before surgery, or had a longer duration of preoperative traction seem to have a higher risk of developing PPNTI.
Mechanism of injury
Common mechanisms of injury for hip procedures are either from the placement of posterior acetabular retractors or from anterior or lateral traction on the femur. The posterior inferior acetabular retractor lies in close proximity to the sciatic nerve; during hip flexion, the nerve can impinge on the acetabular retractor. Traction of up to 40 to 57 kg may be necessary to sufficiently distract the joint for adequate visualization, and it can be very difficult to find the right balance between proper traction and not excessively stretching the nerve to the point of injury. Positions that cause hyperflexion, abduction, and extension of the leg results in stretching of nerves, which results in injury after a prolonged period of time. The evaluation of the function of the sciatic nerve through the intraoperative monitoring of somatosensory cortical evoked potentials has implicated limb positioning and retractor placement in nerve injuries. In those cases, evoked potentials returned to baseline after the repositioning of retractors and/or limbs. Satcher and colleagues used motor evoked potentials in combination with electromyography monitoring during revision THA in 27 consecutive patients and found changes in monitoring parameters associated with leg position, manipulation of acetabular allograft, and insertion of Kirschner (K) wires in the acetabulum, which returned to baseline after extending the hip or repositioning the acetabular allograft or K wires in all cases. In fact, in their study, leg positioning was the most commonly associated intraoperative factor causing changes in monitored parameters.
Another proposed mechanism in total hip replacement is excessive tension caused by lengthening of the extremity. In fact, in cases involving limb lengthening, there was a higher incidence of neuropathy, especially in limbs that were lengthened more than 2.5 cm. However, when Nercessian looked at 1284 cases of THA of which 74% had some limb lengthening involved, he found only 1 case of iatrogenic nerve injury, which was caused by laceration and not traction. There are, unfortunately, no guidelines regarding the amount of lengthening that can be achieved safely.
In the intraoperative monitoring study by Telleria and colleagues, they concluded that maximum traction weight, not the total traction time, is the greatest risk factor for sciatic nerve dysfunction during hip arthroscopy. In their study, the odds of a nerve event increased 4% with every 0.45-kg increase in the traction amount. On the other hand, earlier studies have shown that total traction time was a significant risk factor for developing nerve traction injury, noting an absence of neurologic complications when traction time is limited to less than 2 hours.
As mentioned, the peroneal portion of the sciatic is more commonly involved given its more lateral and superficial nature; in the lateral transtrochanteric approach, it would be difficult to compress the tibial portion without compressing the peroneal portion. This point may account for the fact that when both divisions are injured, the peroneal injury is more severe.
In hip arthroscopy, the joint capsule often requires distraction forces to facilitate the introduction of the scope and instruments into the joint, necessary to visualize the innermost depths of the hip socket, but unfortunately also the mechanism of PPNTI if done for too long.
Injury manifests as paralysis of the hamstring muscles and all the muscles below the knee leading to weak knee flexion and foot drop. There is impairment of sensation below the knee with the exception of the medial aspect of the leg and foot.
Prevention and treatment
Prevention of PPNTI of the sciatic nerve is not only dependent on good operative technique but also more on the need for constant attention to limb positioning, placement of retractors, as well as the weight and duration of retraction. Keeping the amount of traction to less than 50 lb can help reduce the risk of nerve injury. Temporary repositioning of the limbs should be considered during lengthy operations. Traction should be kept to less than 2 hours, or the surgeon should attempt to identify situations when traction can be released or reduced to decrease tension on the nerve. Tellerias’s intraoperative monitoring findings have shown that up to 75% of nerves have shown at least partial recovery after 15 minutes of rest. This rest may involve addressing the peripheral compartment first with minimal traction, removing negative pressure suction seal of the hip before initiating traction, increasing traction slowly with frequent fluoroscopic checks, or even performing capsulotomies. Some have recommended the use of tensiometers during surgery for careful monitoring of the distraction forces and the length of time they are applied.
Given the risk of sciatic nerve injury during hip arthroplasty, some centers now routinely monitor for potential nerve injury with intraoperative monitoring using evoked potentials and free run electromyogram (EMG) to warn surgeons of potential peripheral nerve damage during surgery. However, it remains controversial whether intraoperative nerve monitoring can prevent postoperative sequelae; at least one study found no difference in rates of nerve complications.
In posterior acetabular retraction, the surgeon should try to keep the hip extended whenever possible. When there is extensive acetabular dissection and the use of structural allografts, the position of the sciatic nerve should be assessed carefully because these were associated with sciatic nerve compression. Distraction forces may need to be modified in individual cases given the difference in the inherent laxity because too much force will likely result in hyperelongation of the leg and neuropraxia of the nerves at risk.
Navarro and colleagues came up with and recommended a 5-step surgical approach for reducing the incidence of sciatic nerve palsy: (1) direct finger palpation of the nerve within the posterior scar tissues; (2) limited sciatic neurolysis in difficult cases with extensive scarring; (3) retractor placement guided by the above; (4) final assessment by repeated palpation of the effects of the intraoperative changes in limb position on tension in the nerve and/or compression by a bony or prosthetic prominence; and (5) assessment of tension in the nerve or compression by a bony or prosthetic prominence during trial reduction and range-of-motion testing at the end of the case. In this study, they were able to decrease the incidence of sciatic nerve palsy by 50%. It should be mentioned that although limited sciatic neurolysis of nerves that are tethered or difficult to locate and increased intraoperative attention with palpation of the nerve before and after arthroplasty may help to decrease the prevalence of nerve injury, extra care must be taken because these techniques may also damage the anastomotic blood supply and cause increased scarring.
Others, such as Yacoubian and colleagues, advocate routine visual sciatic nerve identification and tagging in revision hip surgery as one method to potentially reduce the risks of sciatic nerve injury, especially in hip revision given that scarred fibers of the gluteus maximus can easily be mistaken for the sciatic nerve, and relying solely on finger palpation may often be misleading. In the 350 cases in which this surgical technique was implemented, no permanent cases of sciatic nerve injury occurred.
Treatment has mostly been supportive and nonoperative. Physical therapy is prescribed to address motor deficits and prevent joint contractures. Orthoses are prescribed to treat foot drop, allowing clearance during the swing phase and preventing the steppage gait that indicates weak dorsiflexors and also help prevent equinus deformity. Select instances when surgical intervention may be considered include cases of documented limb lengthening, hematoma evacuation, or if there is suspicion of other mechanical causes of neuropathy.
Pudendal nerve injury (S2–S4)
Amarenco and colleagues calculated a total incidence of 2% of pudendal nerve injury secondary to an orthopedic operation. In hip arthroscopy, pudendal nerve injury may occur in as many as 10% of cases. Locker and Beguin reported the first 2 cases of pudendal neurapraxia in hip arthroscopy. Glick and colleagues reported 4 pudendal nerve injuries in a series of 16 arthroscopies. Funke and Munzinger reported 1 pudendal nerve injury in 19 cases. A meta-analysis by Byrd found 10 neurapraxias in 1491 cases.
This nerve is most vulnerable to injury at its distal anatomic course, where it runs upwards toward the pubic symphysis.
Mechanism of injury
Nerve compression is the most likely pathophysiological mechanism. When looking at pudendal neuropathy after femoral nailing, Brumback and colleagues measured the pressure applied on the perineum and found a significant difference in the pressure/operative time ratio between groups with and without pudendal neuropathy (73 kg/h vs 35 kg/h, respectively). They did not find a significant difference, however, in the operative time (168 minutes vs 158 minutes). The suggested mechanism of injury during hip arthroscopy might be compression of the rami during traction against the perineal post.
Initially, there may be perineal and groin pain followed by a sensory deficit. In some cases, this deficit is associated with sexual disorders, such as impotence or anejaculation.
Prevention and treatment
Recommendations for prevention or reducing the risk of pudendal nerve injury include limiting traction to only critical operative moment of operation and the use of good padding on the foot plate and in the perineum. Lindenbaum and colleagues showed that pressure on the perineum can be reduced by using a perineal post with a bigger diameter that provides better force distribution. The recommended diameter of the perineal post is 8 to 10 cm. The ideal position of this perineal post is between the intact lower limb and the external genital organs. Another recommended preventive measure is the ventral decubitus position on the fracture table.
Femoral nerve injury (L2–4)
The femoral nerve is rarely injured after primary hip arthroplasty, (0.04%–2.6% of cases) and in general have been more common in obstetric and pelvic surgeries (most caused by self-retaining retractors ) than in orthopedic procedures. Although considered rare, it is likely underreported because it is frequently self-limiting. The incidence is higher in revision surgery ranging in 1.0% to 7.6% of cases.
After passing beneath the inguinal ligament, the femoral nerve is in close proximity to the femoral head and acetabular rim, the tendon insertion of the vastus intermedius, the psoas tendon, the hip, and the joint capsule. At this level, there is little protection to the femoral nerve from retractors. This fact may explain why there is higher incidence of femoral neuropathy in lateral and anterolateral approaches to the hip joint. The left femoral nerve may be more vulnerable to ischemic injury because of the greater number of branches and richer anastomosis of the deep circumflex iliac artery on the right.
The fascicular architecture of the femoral nerve may also contribute to its low susceptibility to injury because as the anterior division fibers travel separately in a flat fascicle situated ventromedially within the nerve, they may be more exposed to trauma within the pelvis; at the same time, they are less likely to sustain PPNTI caused by orthopedic surgery.
Mechanism of injury
Procedures that require the lower extremity to be positioned in an acutely flexed, abducted, and externally rotated position for long periods can cause traction injury and also compression by angling the femoral nerve beneath the inguinal ligament. Prolonged retraction exposure of the hip, in an anterior or anterolateral approach, is also a mechanism of retractor-induced injury. Improper placement of the anterior acetabular retractor as the tip of the Hohmann retractor is placed near the femoral nerve is a commonly cited mechanism of femoral nerve injury. Simmons and colleagues concluded that all femoral neuropathies in their series were secondary to retractor placement with direct compression of the nerve. In THA, the femoral nerve is most at risk during placement of anterior acetabular retractors when anterior or anterolateral approaches are used. Limb lengthening may endanger the nerve through stretch, although the sciatic rather than the femoral nerve is particularly at risk with this procedure.
The classic presentation is diminished or lost sensation in the distribution of the saphenous nerve (anterior thigh and medial aspect of the leg) with weak hip flexion and loss of extension of the knee. Decreased or absent patellar reflex may also be present.
Prevention and treatment
There should be care taken with positioning patients and limb maneuvering. Exaggerated hip flexion, abduction, and external rotation should be avoided. Temporary repositioning of the limbs should be considered during lengthy operations. In an anterolateral approach, placing the tip of the retractor against the acetabular margin needs to be avoided.
Treatment is partly dictated by the suspected mechanism and severity of injury, and prompt and accurate diagnosis is very important. A compressive hematoma may need drainage and any underlying coagulopathy corrected. Surgical exploration is indicated for injuries that fail to show adequate signs of recovery within about 3 months. This exploration may include repair, grafting, or neurolysis. Nonoperative management should include physical therapy to prevent muscle atrophy and reduce the risk of deep vein thrombosis. A removable brace, such as a knee immobilizer, to hold the knee in extension should be used to prevent contracture and allow safe ambulation.
Following Knee Surgery
Peroneal nerve injury (L4–5 S1–2)
Peroneal nerve injury causing foot drop is a well-known complication of total knee arthroplasty. The incidence varies from 0% to 9.5%. The overall incidence of perioperative injury is estimated to be 0.79%. Again, more than one etiologic factor including PPNTI exist for peroneal nerve injury, and the cause is unknown in most cases. True incidence may be in the higher range because of the presence of atypical peroneal nerve impairments that have different symptoms.
The common peroneal nerve is most susceptible when it courses around the fibular neck and passes through the fibro-osseous opening in the superficial head of the peroneus longus muscle. This opening can be quite tough and can result in the nerve angulating through it at an acute angle. Also, significant fibrous connective tissue secures the nerve to this proximal portion of the fibula, potentially compromising the nerve. The common peroneal nerve may also be more susceptible to traction injuries because it is relatively more fixed between the sciatic notch and the fibular head.
A valgus deformity of the knee joint of at least 10° is a known risk factor for peroneal PPNTI. A preoperative flexion contracture of the knee is another predisposing factor because it increases stretch on the nerve. However, not all studies found that flexion contractures or valgus deformities were significant risk factors for the development of peroneal nerve palsy. Postoperative epidural analgesia may be another significant risk factor for the development of PPNTI of the peroneal nerve. The thought is that epidural analgesia may cause patients to inadvertently rest the limb in a position that directly compresses the peroneal nerve at the fibular head. As mentioned earlier, rheumatoid arthritis is another known risk factor for peroneal nerve PPNTI, although some question if the risk is directly from the effect of rheumatoid arthritis or caused by sequelae of the disease that are themselves risk factors (eg, preoperative valgus deformity and flexion contracture). The length of tourniquet use may be another risk factor, with a total tourniquet time greater than 120 minutes thought to be a significant risk factor for neuropathy.
Mechanism of injury
Intraoperative correction of a valgus deformity of the knee results in traction on the peroneal nerve and PPNTI. The stretch of the nerve and the surrounding soft tissues that occurs during correction of valgus deformity and flexion contracture is thought to result in compromised blood supply to the nerve and subsequent injury. Lateral position is a common risk factor because the nerve is potentially compressed at the fibular head. The nerve is also at risk from compression over the fibular head by an external agent, such as the continuous passive motion machine. The mechanism of injury caused by tourniquet use seems to be caused by both ischemia and mechanical deformation. EMG changes were seen in up to 75% of patients in studies of patients whose surgery was performed with the use of pneumatic tourniquet. What is still uncertain is whether the EMG changes were clinically significant. Other studies have not shown a significant relationship between tourniquet pressure/duration and nerve injury. Nevertheless, the common recommendation is to limit tourniquet use to 2 hours or less because most studies performed to identify the risk factor for peroneal nerve injury after total knee arthroplasty did not find the use of a tourniquet to be a significant factor if the duration was 2 hours or less.
Commonly, patients with motor deficits, such as weakness or even absence of ankle dorsiflexion or extension of the great toe, will result in foot drop or dragging of the toes during ambulation. Sensory manifestations are described along the anterolateral border of the leg and the dorsum of the digits except those supplied by saphenous and sural nerves.
It is important to note that some patients may not present with the typical peroneal nerve palsy symptoms yet still have a peroneal nerve injury. These patients present with mild atypical symptoms, and their only clinical manifestation of peroneal nerve injury may include difficulty with rehabilitation, achieving suboptimal range of motion, or persistent transient symptoms that interfere with their daily activities.
Prevention and treatment
If a PPNTI is suspected, the knee should be repositioned to 20° to 30° of flexion and any constrictive dressing should be changed to relieve traction or possible local compression. This practice is followed with physical therapy and nonsteroidal antiinflammatory drugs/neuroleptics for any pain. Surgical exploration is recommended for any patient not showing either clinical resolution or EMG improvement after 3 months, especially in patients with more severe injuries.